The long term objective of this research is to understand how the T cell repertoire is selected. Two constraints are placed on the ultimate repertoire of mature T cells: no cells should be strongly self-reactive while most or all cells should be self MHC-restricted. This requirements can be met in the context of a 3-part model based on the affinity of binding of an immature T cell's receptor to "self" within the thymus: 1) Thymocytes which react strongly to self MHC and associated self antigens are aborted; there is experimental evidence for this form of "negative selection". 2) Given that T cell receptors are randomly generated the majority of thymocytes will have no affinity for self MHC and are neither positively nor negatively selected but die in situ. 3) The small proportion of thymocytes whose receptors have the "right" affinity (low but real) for self MHC undergo "positive selection" and are exported to the periphery as mature T cells. The present application will address a prediction of positive selection: that mature T cells should bind to cells bearing the class I or class II MHC molecule that was used for their selection. Self-recognition is revealed by the ability of activate alloreactive cytotoxic T lymphocytes to kill bystander targets bearing MHC antigens syngeneic to the CTL but not third-party MHC antigens. This phenomenon is referred to as syngeneic bystander killing. A carefully selected panel of CTL clones and bystander target cells derived from MHC- recombinant mice will be used to determine whether syngeneic bystander killing is restricted to a single MHC-encoded molecule. Monoclonal anti- MHC antibodies will be employed to confirm that syngeneic bystander killing involves receptor mediated recognition of self MHC. That self-recognition is a general characteristic of T cells will be deduced by determining whether Class I and class II MHC-specific, alloreactive MLR blasts as well as class I and class II-restricted, antigen-specific CTL clones exhibit syngeneic bystander killing. Determining if syngeneic bystander killing by CTL derived from allogeneic thymic chimeras is restricted to the MHC expressed in the thymus in which they matured will show that self- recognition is directed by MHC antigens expressed in the thymus. Finally, by determining whether any cell can serve as a syngeneic bystander target, the possible in vivo relevance of activated killer cells which kill syngeneic bystander cells will be addressed.